Learning representation of the enclosing subgraph of node pairs is recognized as an efficient approach for link-oriented prediction tasks in network applications. The core challenge within this subgraph encoding approach is how to effectively distinguish and then properly aggregate the contribution of nodes in the subgraph into a single vector to indicate the relation between the target node pair. In this work, we propose a novel sphere-based subgraph encoding architecture, namely BS-SubGNN, to address the challenge. In detail, we design two key building blocks, including Bicentric Sphere Node Labeling (BSNL) and Bicentric Sphere Subgraph Pooling (BSSP) to assist message passing in BS-SubGNN. BSNL endows each node a label according to the sphere it belongs to in the subgraph to distinguish the contribution of nodes, while BSSP adopts an attention mechanism to aggregate the contribution of nodes in each sphere. Theoretically, we prove that BS-SubGNN can unify existing node distance labeling methods, and yield discriminative node features with less time complexity. We evaluate the performance of BS-SubGNN in link prediction tasks over a variety of network types, including undirected networks, attribute networks, directed networks, and signed directed networks. Our experimental results demonstrate that BS-SubGNN consistently achieves significant performance improvements over the above diverse types of networks. In particular, compared to those methods with a requisite of multi-hop neighborhood information, BS-SubGNN can obtain better performance even when only one-hop neighborhood information of the node pair is utilized.

Efficient long-term flight test task allocation is crucial for aircraft development, yet it poses significant challenges due to the complex, extended timelines and numerous interdependent tasks. Current methods often rely on expert judgment, leading to inefficiencies, delays, and suboptimal planning. This paper addresses these challenges by introducing a reinforcement learning-enhanced two-stage evolutionary algorithm. The algorithm incorporates a mixed-integer nonlinear programming model aimed at minimizing the maximum flight test duration while adhering to multiple constraints. Key innovations include a two-stage evolutionary framework with optimized encoding schemes and crossover operators, and dynamic adjustment of crossover and mutation probabilities using Q-learning. These features enable the algorithm to handle the complexity and unpredictability of long-term flight test planning. Extensive experiments validate the approach, demonstrating its ability to generate high-quality, adaptive flight test plans in real-world, complex scenarios.

This paper proposes a Parallel Fuzzy Nonlinear Active Disturbance Rejection Control (FNLADRC) strategy to improve the precision and robustness of robotic manipulators in machining large, complex components. By decoupling the multi-degree-of-freedom dynamics and integrating Nonlinear ADRC with fuzzy logic, the method adaptively estimates and compensates for disturbances in real time, effectively mitigating machining vibrations and trajectory errors. A parallel control architecture enhances multi-joint coordination, improving adaptability and response speed. Additionally, a fuzzy logic-based tuning mechanism dynamically adjusts control parameters to boost robustness. Simulations on a UR5 robotic arm validate the method’s superior performance in dynamic, uncertain environments, demonstrating FNLADRC as a promising solution for high-precision robotic machining.

As significant sources of energy consumption and carbon emissions, data centers have become a focal point for improving energy efficiency worldwide. To address the challenges of high computational resource demands and limited adaptability of traditional prediction models to complex conditions, this paper proposes an artificial intelligence-enabled predictive energy saving planning based on the Transformer-GRU model for predicting coolant temperature in the liquid cooling system of data centers. By integrating the self-attention mechanism of the Transformer and the time-series prediction strengths of GRU, the model performs correlation analysis and feature extraction of key parameters to achieve high-precision predictions of coolant return temperature. Experimental results demonstrate the model’s superior accuracy compared to traditional prediction models, achieving an MSE of 1.349, RMSE of 1.157, MAPE of 0.0244, and R 2 of 81.07 %, significantly outperforming baseline models such as Transformer-LSTM (MSE = 1.355), Informer (MSE = 1.356), Reformer (MSE = 1.353), DeepAR (MSE = 1.385), LSTM (MSE = 1.351), GRU (MSE = 1.366), and CNN-GRU (MSE = 1.363). The model maintains high predictive accuracy under fluctuating environments and complex cooling conditions, effectively reducing the operational energy consumption of the liquid cooling system. This advancement not only enhances cooling efficiency but also drives data centers toward greater intelligence and sustainability. By leveraging real-time monitoring data and predictive control, the model dynamically optimizes cooling strategies, reducing coolant and energy usage while promoting sustainable resource utilization. Additionally, this study offers implementation insights for high-performance computing environments, laying the groundwork for future research on extending model capabilities and integrating multimodal data.

Efficiently aggregating 4D light field information to achieve accurate semantic segmentation has always faced challenges in capturing long range dependency information (CNN-based) and the memory limitations of quadratic computational complexity (Transformer-based). Recently, the Mamba architecture, which utilizes the state space model (SSM), has achieved high performance under linear complexity in various vision tasks. However, directly applying Mamba to 4D light field scanning will lead to an inherent loss of multi-spatial-angular information. To address the above challenges, we introduce LFSSMam, a novel Light Field Semantic Segmentation architecture based on the selective state space model (Mamba). Firstly, LFSSMam presents an innovative spatial-angular selective scanning mechanism to decouple and scan 4D multi-dimensional light field data. It separately captures the rich spatial context, complementary angular and structural information of light field 2D slices within the state space. In addition, we design an SSM-attention Cross-Fusion Enhance Module to perform preferential scanning and fusion across multi-spatial-angular-modal light field information, adaptively aggregating and enhancing the central view features. Comprehensive experiments on synthetic and real world datasets demonstrate that LFSSMam achieves leading edge SOTA (State-Of-The-Art) performance (with a 6.97% improvement to LF-based methods) while reducing memory and computational complexity. This work provides valuable guidance for the efficient modeling and application of multi-spatial-angular information in light field semantic segmentation. Our code is available at https://github.com/HNU-WQW/LFSSMam.

This paper presents a two-time-scale control method to optimize the energy consumption of high-performance-computing data centers through dynamic frequency scaling of processors, tasks assignment, and cooling supplement. First, the steady and dynamical models of the data center are built, which reflect the computational interactions and thermal relationship among the components of the data center. Next, the energy minimization problem for processing a parallel task is divided into two parts that correspond to the steady thermal model and the dynamic thermal one. Then, the problem is solved in a two-time-scale manner, i.e., the optimization of task assignment and processing frequency is considered in steady thermal environment, and the optimization of cooling supplement is achieved in dynamic thermal environment. Finally, simulations of a real task trace are carried out, which demonstrate that the proposed method can significantly improve energy efficiency while guaranteeing the thermal constraints of the data center.

A model predictive control (MPC) method is presented for the optimizing energy consumption of Internet data centers at the same time maintaining the quality of service (QoS). A dynamical model reflecting computational interactions and thermal relationship between each components of the data center is presented. The model is used to formulate a constrained nonlinear optimal control problem to minimize the energy consumption of both information technology system and cooling technology system. The constraints of this optimal control problem can capture key design requirements, including QoS and thermal constraints for device reliability. Solving this optimal control problem in MPC fashion provides optimal server provisioning, job scheduling, and thermal management techniques that meet specific desired service quality requirements. To reduce the computational delay in solving the constrained nonlinear optimization problem, an approximation algorithm is developed to provide a fast evaluation of complex constrains. The performance of the proposed method is studied through simulations. The results show that significant energy saving can be achieved with the guaranteed throughput of data center.